Knowledge, Attitudes and Practices of Young People in Zimbabwe On

Mapanga et al. BMC Cancer (2019) 19:845 https://doi.org/10.1186/s12885-019-6060-z

RESEARCH ARTICLE Open Access Knowledge, attitudes and practices of young people in Zimbabwe on cervical cancer and HPV, current screening methods and vaccination Witness Mapanga1,2,3* , Brendan Girdler-Brown1 and Elvira Singh1,4,5

Abstract Background: The rise in cervical cancer trends in the past two decades has coincided with the human immunodeficiency virus (HIV) epidemic especially in the sub-Saharan African region. Young people (15 to 24 years old) are associated with many risk factors such as multiple sexual partners, early sexual debut, and high HIV incidences, which increase the chances of developing cervical cancer. The National Cancer Prevention and Control Strategy for Zimbabwe (2014–2018) highlights that no cancer communication strategy focusing on risk factors as primary cancer prevention. Therefore, the study aims to determine the knowledge, attitude and practices of young people in Zimbabwe on cervical cancer, screening, human papillomavirus (HPV) and vaccination. Methods: A cross-sectional survey assessing young people’s knowledge, attitude and practices concerning cervical cancer was conducted in five provinces in Zimbabwe. A total of 751 young people were recruited through a three- stage cluster design from high schools and universities. Knowledge, attitudes and practices were assessed using questions based and adapted from the concepts of the Health Belief Model (HBM) and the Cervical Cancer Measuring tool kit-United Kingdom (UK). Results: Most young people, 87.47% (656/750) claimed to know what the disease called cervical cancer is, with a mean score of 89.98% [95% CI 73.71.11–96.64] between high school and 86.72% [95% CI 83.48–89.40] among university students. There was no significant difference in mean scores between high school and university students (p = 0.676). A risk factor knowledge proficiency score of ≥13 out of 26 was achieved in only 13% of the high school respondents and 14% of the university respondents with a broad range of misconceptions about cervical cancer risk factors in both females and males. There was not much difference on comprehensive knowledge of cervical cancer and its risk factors between female and male students, with the difference in knowledge scores among high school (p = 0.900) and university (p = 0.324) students not statistically significant. In contrast, 43% of respondents heard of cervical cancer screening and prevention, and 47% knew about HPV transmission and prevention. Parents’ educational level, province and smoking, were some of the factors associated with knowledge of and attitude towards cervical among high school and university students. (Continued on next page)

* Correspondence: [email protected] 1School of Health Systems and Public Health, Epidemiology & Biostatistics, University of Pretoria, 5-10 H.W. Snyman Building, Pretoria, South Africa 2Centre for Health Policy, School of Public Health, University of Witwatersrand, Johannesburg, South Africa Full list of author information is available at the end of the article

© The Author(s). 2019 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Mapanga et al. BMC Cancer (2019) 19:845 Page 2 of 14

(Continued from previous page) Conclusion: This study revealed that young people in Zimbabwe have an idea about cervical cancer and the seriousness thereof, but they lack adequate knowledge of risk factors. Cervical cancer education and awareness emphasising causes, risk factors and care-seeking behaviours should be commissioned and strengthen at the community, provincial and national level. Developing a standard cervical cancer primary prevention tool that can be integrated into schools can be a step towards addressing health inequity. Keywords: Knowledge, Attitude, Young people, Cervical cancer, Zimbabwe

Background their mothers had previously been screened for cervical Engaging in risky sexual behaviour and insufficient cancer [9]. This finding suggests that there is some sort knowledge about health issues remain at alarmingly high of passing down of health knowledge within families es- levels among young people aged between 15 to 24 years pecially when parents have utilised health services. How- old in Africa [1]. This was supported by findings that ever, in a country like Zimbabwe where over 80% of showed that condom use among the 15 to 24-year-olds rural women had no previous knowledge about cervical in sub-Saharan Africa (SSA) was only at 57% for young cancer [10], such passing down of knowledge from par- men and 37% for young women, which was below the ents to children is likely not to exist. Besides, with the 95% target advocated by United Nations General Assem- cultural nature in Zimbabwe, young people’s sexuality is- bly Special Session on HIV and AIDS in 2001 [2]. In sues are rarely discussed openly in families [11], making 2015, 17% of young women aged 15–19 years in it interesting to ask some questions on how much young Zimbabwe reported having had sex with a man 10 years people have learnt from their parents about cervical older than themselves in the past 12 months. Also, HIV cancer. prevalence among young people in Zimbabwe increases On the other hand, recent years have witnessed an in- with age, from around 3% in women aged 15–17 years to crease in risky lifestyle behaviour including early onset 14% among the 23–24-year-olds. Among young men, of sexual activity, multiple sexual partners and age-dis- the HIV prevalence rises from about 2.5% among the parate relationships among the 15 to 24 year age group, 15–17-year-olds to around 6% among the 23–24-year- resulting in high HIV incidence and placing young olds [3]. However, only 64% of young women and 47.5% women at risk [3]. HIV incidence for young women be- of young men have ever been tested for HIV [4]. tween the ages of 15–24 years is reported to be twice Sexual behaviour of both men and women is a risk higher as compared to young men of the same age- factor for cervical cancer. Though they do not develop group in Zimbabwe [4]. Thus the need to understand cervical cancer, assessing knowledge, attitude and prac- the knowledge, attitude and practices of young people tices of young men can be vital if a coordinated inclusive towards cervical cancer. strategy towards prevention of cervical cancer is to be There is scant evidence about young people’s know- formulated as advocated by World Health Organisation ledge and understanding of cervical cancer, risk factors, (WHO) in 2009. Also, involving men in cervical cancer screening and HPV vaccination in the developing world. initiatives such as the human papillomavirus (HPV) vac- Do young people know about cervical cancer? What are cination has a cost-benefit relationship that makes it ne- their beliefs and attitude towards cervical cancer risk cessary for them to be incorporated in cervical cancer factors, screening and HPV vaccination? Do they know prevention strategies [5]. Knowledge and awareness of how and where to access cervical cancer services in the cervical cancer and HPV are consistently low across de- country? Therefore, the study aimed to determine the veloping countries and such lack of knowledge provides knowledge, attitude and practices of young people in a challenge to the implementation of cervical cancer Zimbabwe on cervical cancer, screening, HPV and vac- programmes and the new mass HPV vaccination drive cination so as provide relevant information to cervical [6]. Evidence from a developing country among women cancer stakeholders and programme implementers on aged 18 to 44 years old indicated that the majority of how to attract young people in addressing cervical women are unfamiliar with cervical cancer, HPV, vaccin- cancer. ation and screening and that they face several barriers accessing cervical cancer screening services [7, 8]. Methods Sources of information where people get to know Study design, setting and period about cervical cancer are still limited in developing This cross-sectional study took place in six high schools countries. For example, a study suggested that vacci- and five universities in five provinces in Zimbabwe from nated girls were likely to know about cervical cancer if August to November 2017. The selected sites were Mapanga et al. BMC Cancer (2019) 19:845 Page 3 of 14

located in urban and rural settings in each of the five districts comprised the third stage of sampling. High provinces. schools participants who provided consent were recruited for the study through a modified systematic ran- Study population dom sampling of every fifth student. High school class The study population was all young people, 15 to 24 lists were used as sampling frames. There was an auto- years old attending high school or university in matic inclusion in the study for the universities within Zimbabwe. the five selected provinces. If a selected province had more than one university, a random selection of one Sample size determination and sampling method university among the total was carried out (see Add- Power analysis was conducted assuming a power of itional file 1). Purposive sampling was used to select uni- .80 and an alpha of .05. An average response rate of versity participants. This was because it was difficult and 85% was factored-in to adjust for sample sizes. A de- challenging in terms of the logistics to disrupt lectures sign effect of 3 was also factored-in as an adjustment to have all the potential participants in a central place. to the survey sample size, due to the cluster sampling Sex was not considered as a selection criterion for either that was involved. Two distinct sample sizes of 531 high school or university participants. and 447 were calculated for the 15–19 and 20–24- The following five provinces, Mashonaland West, Mid- year-olds, respectively. lands, Masvingo, Manicaland and Harare, were selected The study participants were recruited from six high for inclusion out of the ten provinces in Zimbabwe (see schools and five universities in five of the ten provinces Fig. 1). Harare has more than one university; therefore, a in Zimbabwe. Two separate samples, high school and random selection of one university was carried out as university students, were chosen differently. Three-stage well as a selection of two high schools. cluster sampling was used to select study participants (see Additional file 1). Zimbabwe’s ten provinces were Data collection used as the sampling units for the first stage. The ten Data were collected from August to November 2017 provinces were written on separate paper sheets and using a self-administered questionnaire in the presence these were put in a box where a lottery method was used of the researcher or research assistants. At all levels of to select five provinces. The districts of the selected five data collection and aggregation, there was continuous provinces were used as the sampling units for the second checking of data quality. The questionnaire’s questions stage. The same process was used. All the districts in were based and adapted from the concepts of the Health each of the selected five provinces were placed in a box Belief Model (HBM) and the Cervical Cancer Measuring for lottery selection of one district per previously se- tool kit-United Kingdom (UK) and the content validity lected province. Universities in each of the five selected was established by giving the questions to experts to as- provinces and high schools within the five selected sess their relevance in line with the study objectives.

Fig. 1 Map of the five selected provinces. Source of the editable map https://yourfreetemplates.com/free-zimbabwe-editable-map/1 Mapanga et al. BMC Cancer (2019) 19:845 Page 4 of 14

After validation, the questionnaire was pilot-tested on 40 However, for analytical hypothesis testing and regression randomly selected young people. The questionnaire cov- modelling the clustering inherent in the study design ered a range of issues including: (1) demographics, (2) was taken into account using Stata’s survey (“svy”) cervical cancer knowledge, (3) cervical cancer risk fac- module. tors, (4) HPV knowledge, (5) cervical cancer screening The profile of the respondents was used to identify and vaccination. The HBM dimensions were applied to certain shared or divergent traits. To assess knowledge understand young people’s responses towards cervical on cervical cancer, cervical cancer risk factors, cervical cancer (perceived susceptibility, perceived seriousness of cancer screening and HPV vaccination, frequencies and the disease), their lifestyle concerning cervical cancer percentages were used to express the results. Cronbach’s (perceived benefits), health-seeking nature (perceived alpha was used to measure the internal consistency of barriers) and how their demographic characteristics the Likert questions used to form construct variables. (modifying variables) affect these perceptions. To determine the factors associated with knowledge of The total score on knowledge was calculated by com- cervical cancer, its risk factors, screening and HPV vac- bining the scores of the following three sections: know- cination, logistic regression models were used. Variables ledge of cervical cancer risk factors including HPV with a p-value of 0.25 or under in univariate analyses knowledge and vaccination; knowledge of perceived were unconditionally included in the initial saturated groups at high risk of developing cervical cancer; and backward stepwise regression model [12]. Following knowledge of cervical cancer treatment. The maximum stepwise hierarchical backwards regression modelling, possible score for the knowledge of cervical cancer was explanatory variables were only removed from the 26 and scoring 13 or more was classified as knowing cer- models if the results of a likelihood-ratio (LR) test vical cancer. The total score on positive attitude towards yielded a p-value of greater than 0.283. Results of the as- cervical cancer was calculated by combining the scores sociation were expressed as adjusted odds ratios with of the following three sections: perceived attitude to- 95% confidence intervals. Post-regression tests were car- wards cervical cancer; perceived feelings towards people ried out to assess the goodness of fit of the regression with cervical cancer; and perceived health-seeking be- model as well as the area under the roc curve [12]. Rela- haviour. The maximum possible score for a positive atti- tionship of association between knowledge and attitude tude towards cervical cancer was 17 and scoring 9 and was determined using Chi-squared tests. Significance above were classified as having a positive attitude to- was assumed at a two-sided value of p < 0.05. wards cervical cancer. The practices of young people to- To each participant of the selected high school sample, wards cervical cancer were measured through open- a weight equal to the inverse of the probability of selec- ended questions on their perceived risk of cervical can- tion was calculated and taken into consideration to ob- cer, ways they are preventing cervical cancer and health- tain estimates of population parameters. The weighting seeking behaviours towards screening. process accounted for the sample selection, important since the initial probabilities of selection were not influ- Ethical considerations enced by population sizes of the sampling units. The The researcher and his two research assistants received weight adjustments coincided with known totals of the 3 days of training. The training covered among others, high schools, districts and province populations. study information giving and informed consent process, administration of the questionnaire and general data Results management processes. Ethical permission to conduct We planned to interview a total of 978 participants, 531 the study was obtained from the Ethics Committee high school and 447 university students. Purposive sam- School of Health Systems and Public Health, University pling was used to recruit university students resulting in of Pretoria, ministries of Health and Child Care, Primary 513 completing the interview; that is 66 more students and Secondary Education, and Higher and Tertiary Edu- interviewed than the initially targeted number. However, cation in Zimbabwe, and the Medical Research Council some high school students did not get interviewed (in all of Zimbabwe. Written informed consent from all partici- cases there was no reason given or lack of parental con- pants and parents (of those students under the age of 18 sent). A total of 751 (238 high school and 513 university) years) was sought before data collection. students participated in the study. The response rate among the high school children was thus 238/531 = 44.82%. Data analysis Data were analysed using Stata Software Version 14.0. Socio-demographic characteristics of intended Descriptive statistics for the study samples were calcu- participants lated without any adjustment for the complex sample The majority of the participants were females in both design since the aim was simply to describe the samples. samples. Female students constituted 68.91 and 60.82% Mapanga et al. BMC Cancer (2019) 19:845 Page 5 of 14

of the participants among high school and university were among both females and males as illustrated by re- samples, respectively (see Table 1). The participants’ sponses from the participants: ages ranged from 15 to 21 years old among high schools and from 18 to 24 years among university students. 23-year-old university female student suggested; "Iam Among high school students, those who were 15 no longer using bathing soap on my reproductive (21.43%) and 18 (24.79%) years old, constituted the big- organ because it contributes to the development of gest numbers. Among university participants, 24.76% cervical cancer"; were 20 years old and only 10.72% were 24 years old. Most of the participants resided in high-density areas; Whilst another 21-year-old female university student 55.46% of high school and 50.88% of university students suggested, "having sex with an uncircumcised male (see Table 2). The majority of both high school (92.44%) partner is dangerous and I wish all men will and university (93.96%) students described themselves as answer the call to be circumcised so that women Christians. Half of the university students, 50.49% (259/ will not have to worry about cervical cancer". 513) had ever consumed alcohol as compared to 12.61% (30/230) of high school students. Only 3.70% (19/513) of university students were married with 0.39% (2/513) Factors associated with knowledge of cervical cancer and having been widowed. its risk factors among high school and university students Multiple variable logistic regression modelling was used Knowledge about the disease called cervical cancer to determine the adjusted association between know- Most young people, 87.47% (656/750) claimed to know ledge of cervical cancer and the following factors: age, what the disease called cervical cancer is, with a mean gender, residence, drinking alcohol, smoking, parents’ score of 89.98% [95% CI 73.71.11–96.64] amongst high education and province. Since 92.44% (220/238) of high school and 86.72% [95% CI 83.48–89.40] among univer- school students and 93.96% (482/513) of university stu- sity students. There was no significant difference in dents reported religion to be Christianity, we decided mean scores between high school and university stu- not to include religion in the regression modelling. On dents (p = 0.676). being predictors of knowledge of cervical cancer, most of these socio-demographic characteristics were not sta- Knowledge score for cervical cancer and its risk factors tistically significant. among high school and university students High school students with parents educated up to O- There was not much difference in a comprehensive levels (OR = 2.5; 95% CI = 1.28–4.93) and a qualification knowledge of cervical cancer and its risk factors, based below degree (OR = 3.13; 95% CI = 1.15–8.49), were al- on the calculated overall scores for both high school and most 3 times more likely to have higher knowledge scores university students (see Table 3). Only 12.80% (21/164) about cervical cancer as compared to high school students of high school female students managed a knowledge with parents with a university degree or a primary level score about cervical cancer and its risk factors of 13 and education (see Table 4). In addition, high school students above as compared to 13.51% (10/74) of high school in Mashonaland West (OR = 2.77; 95% CI = 1.60–4.80) male students. However, the difference in knowledge and Midlands (OR = 1.77; 95% CI = 1.06–2.95) provinces scores among high school students was not statistically were 2 to 3 times more likely to have higher knowledge significant (p = 0.900). This trend was also found among scores about cervical cancer as compared to high school university students, with only 1.28% (4/312) of university students in Harare province (see Table 4). female students scoring a knowledge score about cer- Among the university students, those who smoke were vical cancer and its risk factors of 13 and above as com- almost 8 times likely to have higher knowledge scores pared to 2.49% (5/201) of university male students. The about cervical cancer as compared to those who did not difference in cervical cancer knowledge among univer- smoke (OR = 7.80; 95% CI = 1.29–47.21). Also, university sity students was also not statistically significant (p = students in Harare and Mashonaland West were more 0.324). likely to have higher knowledge scores about cervical Overall, only 43.14% (324/751) had ever heard of cer- cancer as compared to the students in the Midlands, vical cancer prevention or screening and 53.0% (398/ Masvingo and Manicaland provinces. However, these 751) did not know about HPV, how it is transmitted or observed differences among provinces were not statisti- prevented. Some of the students indicated that food, cally significant (see Table 5). having sex with an uncircumcised male partner, smoking, and use of detergents such as bathing soaps and hair Cervical cancer attitudes and care-seeking behaviour removers, are some of the factors contributing to the de- Majority of the participants, 94.27% (708/751), acknowl- velopment of cervical cancer. These misconceptions edged that young people should be concerned about Mapanga ta.BCCancer BMC al. et

Table 1 Distribution of gender and age of participants Level of education Frequency Percentage (%)

High school students (2019)19:845 Female 164 68.91 Male 74 31.09 University students Female 312 60.82 Male 201 39.18 High school students Age Harare Manicaland Mash. West Masvingo Midlands Total Frequency Percent Frequency Percent Frequency Percent Frequency Percent Frequency Percent Frequency Percent 15 22 30.99 0 6 14.29 6 13.64 17 43.59 51 21.43 16 16 22.54 0 12 28.57 7 15.91 11 28.21 46 19.33 17 18 25.35 0 12 28.57 7 15.91 4 10.26 41 17.23 18 15 21.13 21 50.00 7 16.67 11 25.00 5 12.82 59 24.79 19 0 19 45.24 4 9.52 13 29.55 2 5.13 38 15.97 20 0 1 2.38 1 2.38 0 0 2 0.84 21 0 1 2.38 0 0 0 1 0.42 Total 71 100.00 42 100.00 42 100.00 44 100.00 39 100.00 238 100.00 University students 18 5 3.82 2 3.28 0 1 0.94 1 0.85 9 1.75 19 32 24.43 19 31.15 6 6.19 3 2.83 15 12.71 75 14.62 20 32 24.43 19 31.15 24 24.74 19 17.92 33 27.97 127 24.76 21 31 23.66 8 13.11 24 24.74 22 20.75 25 21.19 110 21.44 22 15 11.45 5 8.20 19 19.59 19 17.92 22 18.64 80 15.59 23 8 6.11 4 6.56 10 10.31 20 18.87 15 12.71 57 11.11 24 8 6.11 4 6.56 14 14.43 22 20.75 7 5.93 55 10.72 Total 131 100.00 61 100.00 97 100.00> 106 100.00 118 100.00 513 100.00 ae6o 14 of 6 Page Mapanga et al. BMC Cancer (2019) 19:845 Page 7 of 14

Table 2 Other socio-demographic characteristics of participants Table 2 Other socio-demographic characteristics of participants Frequency Percent (Continued) High school students Frequency Percent Province Yes 259 50.49 Harare 71 29.83 Do you smoke Manicaland 42 17.65 No 480 93.57 Mashonaland West 42 17.65 Yes 33 6.43 Masvingo 44 18.49 Paid employment Midlands 39 16.39 No 498 97.08 Residential area yes 15 2.92 High density suburb 132 55.46 Marital status Low density suburb 53 22.27 Single 492 95.91 Rural area 53 22.27 Married 19 3.70 Religion Widowed 2 0.39 Christianity 220 92.44 Traditional 1 0.42 cervical cancer, with a mean score of 90.30% [95% CI = 85.08–92.59] among high school students and 96.20% Apostolic sect 14 5.88 [95% CI = 94.75–97.63] among university students. The Muslim 2 0.84 mean concern for cervical cancer score was not statis- None 1 0.42 tical significance (p = 0.062) between high school and Ever taken alcohol university students. no 208 87.39 Yes 30 12.61 Positive attitude towards cervical cancer scores among high school and university students Do you smoke There was not much difference in a positive attitude to- No 237 99.58 wards cervical cancer, based on the calculated overall Yes 1 0.42 scores for both high school and university students (see Marital status Table 6). Almost half, 48.17% (79/164) of high school fe- Single 238 100.00 male students managed a positive attitude score towards University students cervical cancer of 9 and above as compared to 60.81% (45/ 74) of high school male students. This difference in posi- Province tive attitude scores towards cervical cancer among the Harare 131 25.54 high school students was statistically significant (p = Manicaland 61 11.89 0.018). Among university students, 27.88% (87/312) of Mashonaland West 97 18.91 university female students had a positive attitude score to- Masvingo 106 20.66 wards cervical cancer of 9 and above as compared to Midlands 118 23.00 31.34% (63/201) of university male students. The difference in positive attitude scores among university students Residential area was however not statistically significant (p =0.427). High density suburb 261 50.88 When the respondents were asked on the perceived Low density suburb 184 35.87 risk of them or their girlfriend or wife (in the case of Rural area 68 13.26 male respondents) developing cervical cancer, 45.34% Religion (258/569) indicated no perceived risk. Some of the rea- Christianity 482 93.96 sons that prompted the no perceived risk were as fol- lows: not being an alcohol drinker or smoker, not using Traditional 4 0.78 contraceptive pill, having been circumcised or having a Apostolic sect 13 2.53 circumcised partner, going for regular medical check- Muslim 6 1.17 ups, being faithful to my partner, not HIV positive, by None 8 1.56 praying and not being a commercial sex worker. Ever taken alcohol no 254 49.51 "I am not worried about cervical cancer neither is my girlfriend because I am circumcised and we are Mapanga et al. BMC Cancer (2019) 19:845 Page 8 of 14

Table 3 Cervical cancer knowledge scores among high school Table 4 Factors associated with knowledge of cervical cancer and university students among high school students Cervical cancer Female Male Total Factors associated with knowledge of cervical cancer among high knowledge score school students Total scores (out of 26) Frequency (%) Frequency (%) Frequency (%) High school students* Univariate Adjusted Odds Ratio Cervical cancer knowledge scores for high school students* model (AORs) Main variable Odds p AOR p 95% Conf. 0 27(16.46) 9(12.16) 36(15.13) Ratio Interval 1 18(10.98) 11(14.86) 29(12.18) parents Reference grp 2 5(3.05) 3(4.05) 8(3.36) education (University degree**) 3 4(2.44) 4(5.41) 8(3.36) O-levels 2.86 0.007 2.51 0.017 1.28 4.92 4 10(6.10) 4(5.41) 14(5.88) Qualification 3.01 0.029 3.13 0.033 1.15 8.49 below degree 5 9(5.49) 8(10.81) 17(7.14) A-levels 3.18 0.048 3.01 0.069 0.88 10.25 6 10(6.10) 1(1.35) 11(4.62) No formal 3.20 0.351 3.00 0.406 0.13 68.02 7 18(10.98) 7(9.46) 25(10.50) education 8 12(7.32) 4(5.41) 16(6.72) Province Reference grp (Harare) 9 8(4.88) 3(4.05) 11(5.00) Mashonaland 2.28 0.019 2.56 0.005 1.59 4.80 10 7(4.27) 5(6.76) 12(5.04) West 11 6(3.66) 2(2.70) 8(3.36) Midlands 1.76 0.052 1.64 0.034 1.06 2.95 12 9(5.49) 3(4.05) 12(5.04) Manicaland 1.31 0.280 1.63 0.094 0.89 2.98 13 8(4.88) 4(5.41) 12(5.04) Masvingo 1.24 0.374 1.45 0.127 0.86 2.43 14 4(2.44) 4(5.41) 8(3.36) Residence Reference grp 15 5(3.05) 1(1.35) 6(2.52) (High-density) 16 4(2.44) 1(1.35) 5(2.10) Rural 1.36 0.402 Total (%) 164 (100) 74(100) 238(100) Low-density 0.89 0.565 Cervical cancer knowledge scores for university students** age*** 1.21 0.494 0 48(15.38) 39(19.40) 87(16.96) drinking alcohol 0.72 0.670 1 48(15.38) 33(16.42) 81(15.79) Gender 1.06 0.900 2 38(12.18) 15(7.46) 53(10.33) *Clustering inherent in the study design was taken into account and regression was done using Stata’s survey (“svy”) module for high school students 3 16(5.13) 10(4.98) 26(5.07) **Primary level was combined with University degree since there was no difference between the two levels of education 4 19(6.09) 18(8.96) 37(7.21) ***Age as a continuous variable is liner in relation to the logit (Box-Tidwell 5 16(5.13) 17(8.46) 33(6.43) test p-value = 0.735) Post regression test, using the Pearson’s goodness-of-fit test was carried out 6 26(8.33) 13(6.47) 39(7.60) without factoring the complex sample design- Pearson’s GOF p-value = 0.430 7 20(6.41) 11(5.47) 31(6.04) 8 34(10.90) 15(7.46) 49(9.55) both faithful", wrote 23-year-old male university 9 16(5.13) 8(3.98) 24(4.68) student. 10 17(5.45) 5(2.49) 22(4.29) Whilst a 24-year-old female university student wrote, 11 8(2.56) 10(4.98) 18(3.51) "commercial sex workers and those who drink alcohol 12 2(0.64) 2(1.00) 4(0.78) or use drugs are the ones that are as likely to develop 13 0(0) 1(0.50) 1(0.19) cervical cancer not me since I am a Christian". 14 1(0.32) 0(0) 1(0.19) 15 0(0) 1(0.50) 1(0.19) Some of the respondents (both males and females) in- ‘ 16 0(0) 1(0.50) 1(0.19) dicated that it is solely the responsibility of those who are likely to develop cervical cancer to seek for cervical 17 1(0.32) 0(0) 1(0.19) cancer prevention’. 18 1(0.32) 0(0) 1(0.19) 20 1(0.32) 2(1.00) 3(1.00) "If I am not mistaken, cervical cancer is a women's Total (%) 312(100) 201(100) 513(100) disease, so women should be the ones to be *Chi-2 test, Pearson: Designed-based F(1, 5) = 0.0176, p = 0.900 responsible and take good care of their health", **Fisher’s exact two-tailed p-value = 0.324 suggested 18-year-old male high school student. Mapanga et al. BMC Cancer (2019) 19:845 Page 9 of 14

Table 5 Factors associated with knowledge of cervical cancer among university students Factors associated with knowledge of cervical cancer among university students University students Univariate model Adjusted Odds Ratio (AORs) Final model Variable Odds Ratio p AOR p 95% Conf. Interval Smoker 4.36 0.074 7.80 0.025 1.29 47.21 Residence Reference grp (High-density) Low-density 4.37 0.073 4.25 0.085 0.82 22.02 Rural 1.93 0.593 2.12 0.554 0.18 25.47 Province Reference grp (Harare*) Midlands 0.18 0.113 0.13 0.079 0.01 1.26 Masvingo 0.20 0.137 0.17 0.122 0.02 1.62 Manicaland 0.35 0.332 0.28 0.250 0.03 2.46 age** 0.79 0.300 drinking alcohol 0.48 0.309 Gender 1.96 0.319 *Mashonaland West was combined with Harare since there was no difference between the two provinces **Age as a continuous variable is liner in relation to the logit (Box-Tidwell test p-value = 0.241) Post regression tests were conducted to check the model fit. Area under the ROC curve = 0.70 for the final model; Hosmer and Lemeshow Goodness of fit test p-values = 0.09; 0.12 and 0.17 respectively (8, 10 and 12 groups)

Other respondents believed that women who develop factors; age, gender, residence, drinking alcohol, smok- cervical cancer are of ‘loose morals’ and ‘ignorant’. ing, parents’ education and province. Religion was not included in the regression modelling. Almost no socio- "I am very particular when it comes to my health; I demographic characteristics were statistically associated go for regular check-ups. Women who develop with a positive attitude towards cervical among high cervical cancer are ignorant", wrote a 23-year-old school and university students. High school students female university student. with parents educated up to primary level were 84% more likely not to have a positive attitude towards cer- When asked on what worries them most about cervical cancer (OR = 0.16; 95% CI = 0.06–0.48) as com- vical cancer, the students indicated that cervical cancer pared to high school students with parents with a is associated with dying; the expensive nature of the university degree (see Table 7). treatment; the stigma that society attaches to cervical Among the university students, those in Harare were cancer; failure to conceive; divorce; the suffering and more likely to have higher positive attitude scores to- pain that the patient and their families undergo. wards cervical cancer as compared to the students in the Midlands, Mashonaland West, Masvingo and Manica- "What worries me most about cervical cancer is land provinces. For example, university students in Mid- that the patient will obviously die because the cure lands were 55% more likely not to have a positive is expensive", suggested 15-year-old female high attitude towards cervical cancer as compared to students school student; in Harare (OR = 0.45; 95% CI = 0.25–0.81) and this difference was statistically significant (p = 0.007). University Whilst a 21-year-old male university student wrote, students with parents, who had a qualification below "what worries me most is if my girlfriend or wife is university degree, A-levels, O-levels or no formal educa- to have cervical cancer then I will not have sex and tion, were more likely to have a higher positive attitude she will not be able to bear children for me. That towards cervical cancer as compared to students with can be a recipe for separation". parents with a university degree. However, these observed differences were not statistically significant (see Table 8). Factors associated with a positive attitude towards cervical cancer among high school and university Cervical cancer awareness and need for more information students among high school and university students Multiple variable logistic regression modelling was used A quarter of the young people in this study reported to determine the adjusted association between positive feeling well informed about cervical cancer with mean attitude towards cervical cancer and the following scores of 24.43% [95% CI = 17.14–32.92] and 26.12% Mapanga et al. BMC Cancer (2019) 19:845 Page 10 of 14

Table 6 Positive attitude towards cervical cancer scores among [95% CI 22.32–29.92%] among high school and univer- participants sity students, respectively. There was no significance Attitude towards Female Male Total (p = 0.586) between the two mean scores of feeling well cervical cancer score informed about cervical cancer. However, some of the Total scores (out of 17) Frequency (%) Frequency (%) Frequency (%) young people who claimed to feel well informed about Attitude towards cervical cancer scores for high school students* cervical cancer, also wished for more cervical cancer in- 0 1(0.61) 0(0) 1(0.42) formation, a mean score of 98.09% [95% CI = 92.97– 1 4(2.44) 2(2.70) 6(2.52) 99.50] among high school students and 96.30% [95% CI = 93.66–97.93%] among university students. The 2 36(21.95) 12(16.22) 48(20.17) mean scores were also not significant (p = 0.196) be- 3 13(7.93) 4(5.41) 17(7.14) tween the two groups. 4 4(2.44) 2(2.70) 6(2.52) 5 7(4.27) 0(0) 7(2.94) Factors associated with the need for more cervical cancer 6 9(5.49) 1(1.35) 10(4.20) information among high school and university students 7 6(3.66) 4(5.41) 10(4.20) Logistic regression modelling failed to uncover any meaningful relationships between the measured poten- 8 5(3.05) 4(5.41) 9(3.78) tial explanatory variables and the perceived need for 9 9(5.49) 7(9.46) 16(6.72) more cervical cancer information among high school 10 8(4.88) 9(12.16) 17(7.14) and university students. The logistic regression models 11 9(5.49) 7(9.46) 16(6.72) contained no statistically significant explanatory vari- 12 18(10.98) 7(9.46) 25(10.50) ables and areas under the receiver operating characteris- 13 22(13.41) 8(10.81) 30(12.61) tic (ROC) curve were all less than 0.35. The model F- tests were also non-significant. 14 6(3.66) 5(6.76) 11(4.62) 15 5(3.05) 0(0) 5(2.10) Discussion 16 2(1.22) 1(1.35) 3(1.26) For this investigation, 751 out of the projected 978 indi- 17 0(0) 1(1.35) 1(0.42) viduals were recruited using a three-stage cluster design Total 164(100) 74(100) 238(100) from both high school and universities. This study dem- Attitude towards cervical cancer scores for university students** onstrated that 87% of the young people in Zimbabwe aged 15 to 24 years who participated in this study knew 0 1(0.32) 0(0) 1(0.19) what cervical cancer is. This rate is significantly larger 1 14(4.49) 9(4.48) 23(4.48) than the findings reported in Johannesburg, South Africa 2 88(28.21) 51(25.37) 139(27.10) suggesting that a majority of women 18 to 44 are un- 3 33(10.58) 25(12.44) 58(11.31) familiar with cervical cancer [7], and that reported in 4 30(9.62) 16(7.96) 46(8.97) Zimbabwe showing that over 80% of rural women had 5 22(7.05) 15(7.46) 37(7.21) no previous knowledge of cervical cancer [11]. This finding is even though the National Cancer Prevention and 6 13(4.17) 9(4.48) 22(4.29) Control Strategy for Zimbabwe 2013–2017, which 7 11(3.53) 7(3.48) 18(3.51) encompassed cervical cancer prevention and manage- 8 13(4.17) 6(2.99) 19(3.70) ment, was not fully implemented due to the inadequacy 9 5(1.60) 4(1.99) 9(1.75) of cancer legislation and resource constraints. 10 11(3.53) 4(1.99) 15(2.92) This cross-sectional survey is the first one to be con- 11 11(3.53) 14(6.97) 25(4.87) ducted among young people aged 15 to 24 in Zimbabwe on knowledge, attitude and practices towards cervical 12 12(3.85) 9(4.48) 21(4.09) cancer. This study has shown that there is insufficient 13 15(4.81) 9(4.48) 24(4.68) knowledge about sexual reproductive health including 14 14(4.49) 9(4.48) 23(4.68) cervical cancer among young people. Young people in 15 12(3.85) 5(2.49) 17(3.31) Zimbabwe have a general idea about cervical cancer and 16 4(1.28) 6(2.99) 10(1.95) the seriousness thereof, but they lack adequate know- 17 3(0.96) 3(1.49) 6(1.17) ledge of risk factors and information on where to access cervical cancer services. A risk factor knowledge profi- Total 312(100) 201(100) 513(100) ciency score of ≥13 out of 26 was achieved in only 13% *Chi-2 test, Pearson: Designed-based F(1, 5) = 11.95, p = 0.018 **Fisher’s exact two-tailed p-value = 0.427 of the high school respondents and 14% of the university respondents with a broad range of misconceptions about Mapanga et al. BMC Cancer (2019) 19:845 Page 11 of 14

Table 7 Factors associated with a positive attitude towards cervical cancer among high schools students Factors associated with positive attitude towards cervical cancer among high school students High school students* Univariate model Adjusted Odds Ratio (AORs) Main variable Odds Ratio p AOR p 95% Conf. Interval age** 1.43 0.010 1.36 0.060 0.98 1.89 Gender 1.65 0.018 2.11 0.057 0.97 4.63 Province Reference grp (Harare) Manicaland 1.81 0.027 0.82 0.474 0.42 1.59 Mashonaland West 1.64 0.029 1.67 0.025 1.10 2.54 Masvingo 1.78 0.030 1.32 0.176 0.84 2.08 Midlands 0.77 0.231 0.71 0.154 0.41 1.20 parents education Reference grp (University degree) Primary level 0.14 0.002 0.16 0.007 0.06 0.48 No formal education 0.24 0.158 0.15 0.225 0.01 5.05 A-levels 0.58 0.355 0.75 0.663 0.15 3.80 Qualification below degree 1.19 0.709 1.32 0.561 0.42 4.14 O-levels 0.91 0.783 0.91 0.815 0.34 2.42 drinking alcohol 0.76 0.566 Residence Reference grp (High density) Low density 1.17 0.633 Rural 0.93 0.813 *Clustering inherent in the study design was taken into account and regression was done using Stata’s survey (“svy”) module for high school students **Age as a continuous variable is liner in relation to the logit (Box-Tidwell test p-value = 0.512) Post regression test, using the Pearson’s goodness-of-fit test was carried out without factoring the complex sample design- Pearson’s GOF p-value = 0.182 cervical cancer risk factors in both females and males. where it is offered. There was wide ignorance among the Having heard about cervical cancer and considering it to respondents on cervical cancer risk factors, with a lot of be serious does not necessarily correspond to young myths and misconceptions. These findings point to the people having a correct understating of the disease and need for a cervical cancer primary communication strat- its associated risk factors. There were no significant dif- egy or tool targeting young people to improve their ferences in knowledge of cervical cancer and its risk fac- health education. Besides, considering the long incuba- tors between high school and university students or tion period for someone to develop cervical cancer, it rural and urban students and this might indicate lack of might be argued that targeting young people in cervical cancer education or awareness at a national level. Over cancer prevention, by screening from the age of 15 and 85% of both high school and university students did not after sexual debut, in light of HIV incidence among have enough knowledge about cervical cancer or its risk them, might be beneficial and cost-effective shortly. factors. This lack of knowledge is similar to what was Zimbabwe does not have a cancer primary prevention found among a study that assessed women’s knowledge, strategy that focuses on cancer risk factors [13]. The attitude and practices towards cervical cancer screening HBM and socio-ecological model (SEM) believe that in- in Zimbabwe [10]. creasing knowledge and awareness of a disease, in this The current cervical cancer screening strategies espe- case, cervical cancer, may play a role in improving cially in Zimbabwe target women older than 21 years old healthcare-seeking behaviour among young people to- [13] but this study has shown that the knowledge of the wards cervical cancer prevention services [15–17]. Based screening services and their availability is very low even on the HBM and SEM, lack of information and know- among young women between the ages of 21 to 24 years. ledge about cervical cancer, especially as found in this This finding was similar to what was found in study, are some of the reasons that contribute to under- Cambodia, where women aged 20–69 years had a low utilisation of screening services, late diagnosis of the awareness of cervical cancer screening services, which condition, and a high mortality rate [18, 19]. Young led to underutilisation of these services [14]. Generally, people in this study (including the young men’s percep- young people were unable to mention cervical cancer tion towards their girlfriends and wives) did not feel sus- risk factors or know about cervical cancer screening and ceptible to cervical cancer because of various reasons Mapanga et al. BMC Cancer (2019) 19:845 Page 12 of 14

Table 8 Factors associated with a positive attitude towards cervical cancer among university students Factors associated with positive attitude towards cervical cancer among university students University students Univariate model Adjusted Odds Ratio (AORs) Variable Odds Ratio p AOR p 95% Conf. Interval Province Reference grp (Harare) Midlands 0.46 0.007 0.45 0.007 0.25 0.81 Mashonaland West 0.65 0.148 0.65 0.142 0.36 1.16 Masvingo 0.81 0.443 0.80 0.428 0.46 1.39 Manicaland 0.87 0.676 0.92 0.808 0.48 1.78 parents education Reference grp (University degree) Qualification below degree 1.59 0.057 1.65 0.054 0.99 2.64 No formal education 2.01 0.126 1.94 0.151 0.79 4.82 A-levels 1.45 0.346 1.52 0.299 0.69 3.33 Primary level 0.65 0.582 0.66 0.604 0.14 3.16 O-levels 1.16 0.593 1.19 0.538 0.68 2.10 smoker 0.63 0.298 Gender 1.18 0.401 drinking alcohol 1.09 0.660 residence Reference grp (High-density) Low-density 1.06 0.792 Rural 1.03 0.913 age** 1.00 0.948 **Age as a continuous variable is liner in relation to the logit (Box-Tidwell test p-value = 0.635) Area under the ROC curve = 0.61 for the final model; Hosmer and Lemeshow Goodness of fit test p-values = 0.34; 0.51 and 0.30 respectively (8, 10 and 12 groups) ranging from being faithful, to having being circumcised 24]. Conservatively, even when parents are informed or and not using detergents that are believed to cause can- educated, the culture of sharing their knowledge with cer. This belief of lack of susceptibility towards cervical their children is not often practiced. The findings of this cancer by young people is almost similar to what was study have shown that parents educational level does found in other studies among women [20, 21]. Also, not correlate with young people’s knowledge of cervical there is a lack of cervical cancer prevention policies even cancer, with those with parents with university degrees in light of the newly launched Mass HPV Vaccination having low knowledge as compared to some with par- Programme for young girls in Zimbabwe. These legisla- ents who have secondary education. This also means tion and policy challenges might be contributing to the parental awareness and knowledge might be low as re- fragmentation of cervical cancer service provision and ported in other countries [25]. failure to prevent ‘silo’ operation among different partners within the sexual reproductive health management Study limitations space. Such legislation and policy challenges have been This study had several limitations. The convenient sam- cited as barriers in most developing countries and solv- pling, which was used for university students, made it ing them requires information as reported in this study’s difficult to generalise their sentiments for the whole findings [22]. country. Also, those who participated in this study might Lastly, Zimbabwe remains both a patriarchal and con- have been different from those who could not participate servative society and most men continue to make deci- because they were attending lectures or writing exami- sions on the health of women, from providing for money nations. The research findings might have underesti- for hospital fees to deciding if women must seek medical mated the extent of lack of knowledge, attitude and attention. This study highlights the ignorance that young practices of young people towards cervical cancer due to people, especially young men, have on cervical cancer the non-response rate of high school students. The re- and its screening. Men continue to be sidelined in health sponse rate in the high school students was almost 45% prevention programmes and this has fueled their passive and the inability to obtain parental consent accounted nature towards health-seeking behaviour, as some re- for 6% of this low response. The 55% non-response rate sponses from young men in this study have shown [23, among high school students may reflect a decrease in Mapanga et al. BMC Cancer (2019) 19:845 Page 13 of 14

statistical power and this might potentially have under- Authors’ contributions mined the generalisability of the reported results to the WM designed the study and led the entire field survey; WM performed data cleaning and statistical analysis with BGB and ES providing valuable larger target audience. The university cohort was over- guidance and expertise. All authors read and approved the final manuscript. sampled with 531 of a projected 513 sample size goal. This also might have introduced some bias in overesti- Authors’ information WM is a Cancer Clinical Epidemiologist, and Health Systems and Policy mating the reported results among university students. Researcher and this survey was part of his PhD research project. BGB is an extraordinary professor at the School of Health Systems and Public Health, University of Pretoria. ES is a Public Health Medicine Specialist and Head of Conclusion the South African National Cancer Registry. In as much as young people in Zimbabwe have an idea Funding about cervical cancer and the seriousness thereof; they The Association of African Universities (AAU) and the University of Pretoria lack adequate knowledge on risk factors as indicated by provided the financial support for this survey. The two funding bodies did the broad range of misconceptions. Cervical cancer edu- not play any role or participate in the design of the study, data collection, and analysis, interpretation of data and writing of the manuscript. cation and awareness emphasising causes, risk factors and care-seeking behaviours should be commissioned Availability of data and materials and strengthened at the community, provincial and na- The authors declare that all data supporting the findings of this study are available within the article. The dataset was generated and analysed during tional level. Developing a standard cervical cancer pri- the current study. mary prevention tool that can be integrated into schools can be a step towards addressing this health inequity. Ethics approval and consent to participate This survey was reviewed by both the Research Ethics Committee, Faculty of This standard cervical cancer primary prevention tool Health Science, University of Pretoria (Ethics Reference Number: 146/2016) should be guided by the WHO-coordinated strategy to- and the Medical Research Council of Zimbabwe (MRCZ/A/2135). Written wards the prevention of cervical cancer, which sought to informed consent from all participants and parents (of those students under the age of 18 years) was sought before data collection. include both men and women in cervical cancer initiatives [26]. Involving men in cervical cancer prevention Consent for publication strategies is key to reinforcing the drive to enhance the Written consent was obtained from all the participants for their responses to disease’s management and this has a cost-benefit rela- be published in this study. tionship [27]. Zimbabwe can develop and ring-fence her Competing interests cervical cancer prevention and management plans by de- The authors declare that they have no competing interests. veloping a policy in line with WHO coordinated strat- Author details egy. The policy might need to incorporate primary and 1School of Health Systems and Public Health, Epidemiology & Biostatistics, community cervical cancer prevention and management University of Pretoria, 5-10 H.W. Snyman Building, Pretoria, South Africa. 2 initiatives pool together all public and private cervical Centre for Health Policy, School of Public Health, University of Witwatersrand, Johannesburg, South Africa. 3Harare, Zimbabwe. 4Cancer cancer providers and support a financing system that Epidemiology Research Group, National Cancer Registry, National Health can provide access to affordable, quality cervical cancer Laboratory Service, Johannesburg, South Africa. 5Community Medicine Unit, services. School of Public Health, University of Witwatersrand, Johannesburg, South Africa.

Received: 15 March 2019 Accepted: 19 August 2019 Additional file

References Additional file 1: Three-stage cluster sampling of participants. (DOCX 1. Schaefer R, Gregson S, Eaton J, Mugurungi O, Rhead R, Takaruza A, Maswera 118 kb) R, Nyamukapa C. Age-disparate relationships and HIV incidence in adolescent girls and young women: evidence from Zimbabwe. AIDS. 2017; 31(10):1461–70. Abbreviations 2. Shisana O, Rehle T, Simbayi LC, Zuma K, Jooste S, Zungu N, et al. South AAU: Association of African Universities; HBM: Health Belief Model; African national HIV prevalence, incidence and behaviour survey, 2012. Cape HIV: Human immunodeficiency virus; HPV: Human papillomavirus; Town: Human Sciences Research Council; 2014. LR: Likelihood-ratio; MRCZ: Medical Research Council of Zimbabwe; 3. Zimbabwe National Statistics Agency (2016) “Zimbabwe Demographic and ROC: Receiver operating characteristic; SEM: Socio-ecological model; Health Survey 2015”. SSA: sub-Saharan Africa; UK: United Kingdom; WHO: World Health 4. Zimbabwe National Statistics Agency. Zimbabwe Demographic and Health Organisation Survey 2015: Key Indicators; 2015. p. 39–40. 5. Taira AV, Neukermans CP, Sanders GD. Evaluating human papillomavirus vaccination programs. J Emerg Infect Dis. 2004;10(11):1915–23. Acknowledgements 6. Perlman S, Wamai RG, Bain PA, Welty T, Welty E, Ogembo JG. Knowledge We are very grateful to the following Zimbabwean ministries: Health and and awareness of HPV vaccine and acceptability to vaccinate in sub- Child Care, Primary and Secondary Education, and Higher and Tertiary Saharan Africa: a systematic review. PLoS One. 2014;9(3):e90912. Education, who approved and facilitated our survey in the high schools and 7. Francis SA, Nelson J, Liverpool J, Soogan S, Mofammere N, Thorpe RJ Jr. universities. We also are thankful to all the young people who agreed to Examining attitudes and knowledge about HPV and cervical cancer risk answer our interviews and to the two research assistants, Caroline among female clinic attendees in Johannesburg. Vaccine. Chiumburu and Tatenda Mudehwe, for their tireless fieldwork. 2010;28(50):8026–32. Mapanga et al. BMC Cancer (2019) 19:845 Page 14 of 14

8. Francis SA, Battle-Fisher M, Liverpool J, Hipple L, Mosavel M, Soogan S, Mofammere N. A qualitative analysis of south African women's knowledge, attitudes, and beliefs about HPV and cervical cancer prevention, vaccine awareness and acceptance, and maternal-child communication about sexual health. Vaccine. 2011;29(47):8760–5. 9. Mollers M, Lubbers K, Spoelstra SK, Weijmar-Schultz WCM, Daemen T, Westra TA, van der Sande MAB, Nijman HM, de Melkar HE, Tami A. Equity in human papillomavirus vaccination uptake?: sexual behaviour, knowledge and demographics in a cross-sectional study in (un) vaccinated girls in the Netherlands. BMC Public Health. 2014;14:288. 10. Mupepi SC, Sampselle CM, Johnson TRB. Knowledge, attitudes, and demographic factors influencing cervical cancer screening behaviour of Zimbabwean women. J Women’s Health. 2011;20(6):943–52. 11. Amnesty International. “Lost without knowledge” barriers to sexual and reproductive health information in Zimbabwe: Amnesty International; 2018. Accessed on 13 January 2019 https://www.amnesty.org/download/ Documents/AFR4677002018ENGLISH.PDF 12. Vittinghoff E, Glidden DV, Shiboski SC, McCulloch CE. Regression methods in biostatistics - Linear, logistic, survival, and repeated measures models. (Chapter 5.5.3.). New York: Springer-Verlag; 2004. p. 340. 13. Kuguyo O, Matimba A, Tsikai N, et al. Cervical cancer in Zimbabwe: a situation analysis. Pan Afr Med J. 2017;27:215. 14. Touch S, Oh JK. Knowledge, attitudes, and practices toward cervical cancer prevention among women in Kampong Speu Province. BMC Cancer. 2018; 18(1):294. 15. Aldohaian AI, Alshammari SA, Arafah DM. Using the health belief model to assess beliefs and behaviours regarding cervical cancer screening among Saudi women: a cross-sectional observational study. BMC Womens Health. 2019;19(1):6. 16. Yakout SM, Moawed S, Gemeay EM. Cervical Cancer and screening test (PAP test): knowledge and beliefs of Egyptian women. Am J Nurs Sci. 2016;5:175– 84. 17. Nyambe A, Van Hal G, Kampen JK. Screening and vaccination as determined by the social ecological model and the theory of triadic influence: a systematic review. BMC Public Health. 2016;16(1):1166. 18. Daley E, Alio A, Anstey EH, Chandler R, Dyer K, Helmy H. Examining barriers to cervical Cancer screening and treatment in Florida through a socio- ecological lens. J Community Health. 2011;36:121–31. 19. Mehraban SSZ, Namdar A, Naghizadeh MM. Assessment of preventive behavior for cervical Cancer with the health belief model. Asian Pac J Cancer Prev. 2018;19(8):2155–63. 20. Allahverdipour H, Emami A. Perceptions of cervical cancer threat, benefits, and barriers of Papanicolaou smear screening for women in Iran. Women Health. 2010;47(3):23–37. 21. Baranoski A, Horsburgh C, Cupples L, Aschengrau A, Stier E. Risk factors for nonadherence with pap testing in HIV-infected women. J Women's Health. 2011;20(00):1–9. 22. Stewart, B. and Wild, C.P. (eds.), International Agency for Research on Cancer, WHO. (2014) World Cancer Report 2014 [Online]. Available from: http://publichealthwell.ie/node/725845 [Accessed: 21st May 2019]. 23. Botha MH, Richter KL. Cervical cancer prevention in South Africa: HPV vaccination and screening both essential to achieve and maintain a reduction in incidence. S Afr Med J. 2015;105(1):33–5. 24. Kim HW, Kim DH, Kim Y. Men's awareness of cervical cancer: a qualitative study. BMC Womens Health. 2018;18:155. 25. Wisk LE, Allchin A, Witt WP. Disparities in human papillomavirus vaccine awareness among U.S. parents of preadolescents and adolescents. Sex Transm Dis. 2014;41(2):117–22. 26. Coleman JS, Cespedes MS, Cu-Uvin S, Kosgei RJ, Maloba M, Anderson J, et al. An insight into cervical Cancer screening and treatment capacity in sub Saharan Africa. J Low Genit Tract Dis. 2016;20(1):31–7. 27. World Health Organisation. Comprehensive Cervical Cancer Control: A guide to essential practice. 2nd edition. Australia: World Health Organisation; 2014. ISBN 978 92 4 1548953.

Publisher’sNote Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.